Tobacco Excise Taxation in South Africa - · PDF fileannounced that it would accelerate the...

10

World Health Organization


Corné van Walbeek

University of Cape Town, South Africa

Tobacco Excise Taxation in South Africa

2


Tobacco Free Initiative Headquarters would like to thank the Regional Offices

for their contribution to this project.

WHO Regional Office for Africa (AFRO)

Cite du Djoue

Boîte postale 6

Brazzaville

Congo

Telephone: +(1-321) 95 39 100/+242 839100

WHO Regional Office for the Americas / Pan

American Health Organization (AMRO/PAHO)

525, 23rd Street, N.W.

Washington, DC 20037

U.S.A.

Telephone: +1 (202) 974-3000

WHO Regional Office for the Eastern

Mediterranean (EMRO)

WHO Post Office

Abdul Razzak Al Sanhouri Street, (opposite Children’s

Library)

Nasr City, Cairo 11371

Egypt

Telephone: +202 670 2535

WHO Regional Office for Europe (EURO)

8, Scherfigsvej

DK-2100 Copenhagen

Denmark

Telephone: +(45) 39 17 17 17

WHO Regional Office for South-East Asia (SEARO)

World Health House, Indraprastha Estate

Mahatma Gandhi Road

New Delhi 110002

India

Telephone: +(91) 11 337 0804 or 11 337 8805

WHO Regional Office for the Western Pacific

(WPRO)

P.O. Box 2932

1000 Manila

Philippines

Telephone: (00632) 528.80.01

3


2


Tobacco Free Initiative Headquarters would like to thank the Regional Offices

for their contribution to this project.

WHO Regional Office for Africa (AFRO)

Cite du Djoue

Boîte postale 6

Brazzaville

Congo

Telephone: +(1-321) 95 39 100/+242 839100

WHO Regional Office for the Americas / Pan

American Health Organization (AMRO/PAHO)

525, 23rd Street, N.W.

Washington, DC 20037

U.S.A.

Telephone: +1 (202) 974-3000

WHO Regional Office for the Eastern

Mediterranean (EMRO)

WHO Post Office

Abdul Razzak Al Sanhouri Street, (opposite Children’s

Library)

Nasr City, Cairo 11371

Egypt

Telephone: +202 670 2535

WHO Regional Office for Europe (EURO)

8, Scherfigsvej

DK-2100 Copenhagen

Denmark

Telephone: +(45) 39 17 17 17

WHO Regional Office for South-East Asia (SEARO)

World Health House, Indraprastha Estate

Mahatma Gandhi Road

New Delhi 110002

India

Telephone: +(91) 11 337 0804 or 11 337 8805

WHO Regional Office for the Western Pacific

(WPRO)

P.O. Box 2932

1000 Manila

Philippines

Telephone: (00632) 528.80.01

3


Introduction

The past ten years have witnessed a major turnabout in

government policy on tobacco control in South Africa.

Within a relatively short time, government policy has

changed from complete apathy to one where the tobacco

control measures are regarded as some of the most pro-

gressive in the world.

South Africa’s tobacco control policy rests on two impor-

tant pillars: legislation and excise tax increases. In 1999

the government passed legislation that banned tobacco

advertising and sponsorship, prohibited smoking in all pub-

lic places (including workplaces), and banned the sale of

tobacco to minors. This legislation was an amendment to

an act passed in 1993 that prohibited smoking on public

transport and introduced health warnings for the first time.

As well as increasing the implicit costs of smoking, the

legislation prohibiting smoking in public and work places

represents a clear transfer of property rights from smok-

ers to non-smokers. Whereas previously smokers enjoyed

the right to pollute the air, the legislation unambiguously

assigns non-smokers the right to unpolluted air. Although

the direct impact of the legislation on tobacco consump-

tion is still unclear, the legislation has continued the trend

of deglamorising smoking in South Africa. As a result,

smoking is no longer regarded as socially acceptable by

large sections of the population.

In the past decade the government has substantially

increased the excise tax on tobacco products for health

reasons. Since 1994 the nominal tax on cigarettes has

increased by nearly 25 per cent each year. Econometric

evidence indicates that the resulting price increases have

had a significant impact on cigarette consumption. The

aim of this paper is to investigate tax increases in some

detail.

Description of the Intervention

During the 1970s and 1980s tobacco control was not on

the public agenda. The tobacco industry used its cordial

relations with the government to prevent any measures

that would harm the industry. On tobacco issues, the

government regularly consulted the industry. For exam-

ple, before the budget was presented to Parliament

the tobacco industry was consulted about possible tax

increases. Not surprisingly, the tax increases were generally

very modest. In fact, between 1970 and the early 1990s

the real (i.e. inflation adjusted) excise tax on cigarettes

decreased by 70 per cent. This rapid decrease occurred

despite calls by the medical community and the Ministry

of Health to increase the excise tax.

In 1994 the African National Congress became the domi-

nant party in the Government of National Unity after the

first democratic elections. In the early 1990s the outgoing

government had started introducing some tobacco con-

trol measures, in the form of legislation mandating health

warnings, and increases in the excise tax. In 1993 the ANC

announced that it would accelerate the tobacco control

measures if it came to power.

The new government made its intentions clear at the

reading of the Budget in June 1994, when the Minister of

Finance announced that the government would increase

the tax on tobacco products to 50 per cent of the retail

price.1 At that point, excise taxes amounted to 21 per cent

of the retail price and the total tax burden (i.e. including

sales tax) was 32 per cent of the retail price. However,

after being pressurised by the industry, the government

opted for a slower phasing in of the adjustment. While

the phasing in approach was a temporary setback for the

tobacco control lobby, the government kept to its prom-

ise, increasing the excise tax by substantially more than

the inflation rate at subsequent readings of the Budget. In

1997 the Minister of Finance announced that the 50 per

cent target had been achieved. Subsequent tax increases

were aimed at keeping the tax percentage at that level.

Some trends regarding tobacco taxation in South Africa

are shown in Table 1. Column (f) illustrates the rapid

decrease in real excise tax between 1970 and the early

1990s, followed by a sharp increase subsequently. A

recent study has shown that, in the past decade, South

Africa has had the third highest percentage change in

tobacco taxes (after Korea and France) amongst 90 coun-

tries. It is interesting that, despite the industry’s protes-

1 In South Africa, as in many countries, the excise tax is

levied as a specific tax, i.e. a certain amount per pack of

cigarettes. Unless the tax is adjusted regularly, inflation

will erode the tax. This is exactly what happened in South

Africa during the 1970s and 1980s.

However, even though the excise tax is technically a spe-

cific tax, the government’s policy of setting the tax at 50

per cent of the retail price has turned it into a de facto ad

valorem tax.

4


5


Year Cons, millions of packs

Price, Nominal,

Cents per pack

Price, Real, 1995 base,

Cents per pack

Excise tax, Nominal,

Cents per pack

Excise tax, Real, 1995 base,

Cents per pack

Excise tax as % of price

Total tax as % of price

Industry price, Real, 1995 base,

Cents per pack

Excise revenue, Real, 1995 base,

R millions

(a) (b) (c) (d) (e) (f) (g) (h) (i) (j)

1961 517 19.1 449 9.1 214 47.6% 47.6% 235 1106

1965 608 19.4 417 9.1 196 46.9% 46.9% 222 1189

1970 783 22.1 405 11.1 203 50.2% 50.2% 202 1593

1975 1048 31.8 373 14.6 171 45.9% 45.9% 202 1795

1980 1283 49 328 20.1 134 41.0% 44.9% 181 1725

1981 1443 53 308 20.1 117 37.9% 41.8% 179 1684

1982 1632 62 314 21.1 107 34.0% 39.1% 191 1745

1983 1551 66 298 24.1 109 36.5% 42.2% 172 1686

1984 1570 74 299 24.6 99 33.2% 41.0% 176 1560

1985 1571 84 292 26.1 91 31.1% 41.4% 171 1425

1986 1591 94 276 26.1 77 27.8% 38.5% 170 1217

1987 1671 109 275 26.1 66 23.9% 34.7% 180 1101

1988 1795 122 273 27.1 61 22.2% 32.9% 183 1089

1989 1809 138 269 30.6 60 22.2% 33.5% 179 1079

1990 1868 165 281 33.1 56 20.1% 31.6% 193 1055

1991 1927 171 253 37.6 56 22.0% 32.9% 170 1072

1992 1900 222 288 44.6 58 20.1% 29.2% 204 1100

1993 1802 255 302 53.2 63 20.9% 31.3% 204 1135

1994 1769 284 309 60.5 66 21.3% 33.6% 205 1162

1995 1708 348 348 75.3 75 21.6% 33.9% 230 1287

1996 1690 387 360 92.0 86 23.8% 36.1% 230 1447

1997 1577 497 426 117.5 101 23.6% 35.9% 273 1588

1998 1495 608 487 169.5 136 27.9% 40.2% 292 2032

1999 1422 730 558 214.3 164 29.3% 41.6% 325 2332

2000 1333 803 582 254.5 184 31.7% 44.0% 326 2453

2001b 1272 889 608 291.5 199 32.8% 45.1% 334 2540

a Sources: Auditor-General, Statistics South Africa (previously Central Statistical Services and Department of Statistics), Budget Review, Tobacco Board.

b Preliminary figures.

Table 1

Trends in cigarette prices, taxes and consumptiona

4


5


tations about the “unreasonableness” of the excise tax

increases, the real excise tax in 2001 is no higher than the

level of the 1960s.2 Total tax as a percentage of the retail

price follows a similar trend, as indicated in column (h).

The tax proportion decreased from 50 per cent in 1970 to

30 per cent in 1992, after which it rose to 45 per cent in

2001.3 Despite the excise tax increases, the tax proportion

of South African cigarettes, compared to many Western

countries, is still low.

The real retail price of cigarettes has more than doubled

over the past decade, as is shown in column (d). This

means that cigarettes, in comparison to a basket of other

goods and services, have become very expensive. In fact,

of all commodities surveyed by the South African statistical

Financial year Cigarettes Cigarette

tobacco

Pipe tobacco Cigars Inflation rate

1990/1 11 11 10 13 12.0

1991/2 14 14 11 14 11.4

1992/3 8 8 4 5 8.3

1993/4 9 19 2 2 9.7

1994/5 25 29 25 30 9.0

1995/6 24 27 25 28 8.7

1996/7 18 20 18 19 7.4

1997/8 52 56 52 53 8.6

1998/9 29 31 29 29 6.9

1999/00 20 85 166 3669 5.2

2000/1 16 40 56 74 5.4

2001/2 12 12 20 17 5.7

Average 1990/1

– 1993/4

11 13 7 9 10.4

Average 1994/5

– 2001/2

25 38 49 490 7.1

Average 1990/1

– 2001/2

20 29 35 330 8.2

a Source: Budget Review

2 However, during the 1960s cigarettes, like all goods, were

not subject to sales tax. The imposition of sales tax, since

the late 1970s has increased the effective tax burden

above the level of the 1960s.

3 Since 1997 the Ministry of Finance has claimed that it has

achieved the 50 per cent tax target. This is more illusion-

ary than real. When the Ministry calculates the tax inci-

dence percentage, the denominator they use is the retail

price before the tax increase. This is unrealistic. The tax

increase causes the retail price to increase, with the result

that the denominator increases. So, ex post, the total tax

percentage is much lower than the claimed 50 per cent, as

is illustrated by column (h) of Table 1.

Table 2

Nominal percentage changes in the excise tax on various tobacco productsa

6


7


authorities, cigarettes have been subject to the largest price

increases over this period. This was a dramatic reversal

of the previous 20 years’ trend, since between 1970 and

the early 1990s the real price of cigarettes had fallen by

a third. At purchasing power parity, the price of South

African cigarettes is currently comparable to those of many

European countries and Japan.

In South Africa cigarettes represent more than 90 per cent

of tobacco sales. Some of the poorer sections of society

buy pipe and cigarette tobacco and roll their own ciga-

rettes. However, despite the large increases in the real

price of cigarettes, the non-cigarette tobacco segment has

remained small.

In order to be effective, tobacco excise tax increases

should not create incentives for people to shift their

tobacco consumption from one form to another. The tax

increases should thus be similar for the various tobacco

products. In Table 2 the percentage changes in the excise

tax for the four excisable tobacco products are shown.

In most years the tax increases on potential substitutes

to cigarettes (i.e. pipe and cigarette tobacco) have been

similar to that of cigarettes, but in some years, notably

1999 and 2000, the tax increases have been substantially

greater. This suggests that the government was aware of

the possible substitutability of tobacco products, and did

not want to create an incentive for consumers to switch to

substitutes.

An important omission is snuff, which is not taxed at all.

According to the National Council Against Smoking, in

South Africa, almost as many women use snuff as smoke

cigarettes. Despite tobacco control groups lobbying for a

tax on snuff, no tax has been imposed to date.

In South Africa the excise tax on cigars has traditionally

been very low. However, this changed dramatically in

1999 when the excise tax was increased nearly forty-fold.

This was followed by another 74 per cent increase in the

excise tax in 2000. The Minister of Finance claims that

the large increases in the tax on cigars were necessary to

bring them into line with the tax on cigarettes. However,

the fact that few people in South Africa smoke cigars on

a regular basis, and that they are regarded as luxury and

“special events” items, suggests that the primary aim of

increasing the tax on cigars was to increase government

revenue.

Implementation

In contrast to tobacco control legislation, which has to go

through a lengthy parliamentary process, it is very easy to

increase the excise tax on tobacco products. Even before

1994, the Minister of Finance announced increases in the

tobacco excise tax at the annual reading of the Budget.

However, as pointed out, the increases were small and

usually less than the inflation rate.

In South Africa a vocal tobacco control lobby, led by the

Medical Research Council (MRC) and the National Council

Against Smoking (NCAS), had been arguing for significant

tax increases since the 1970s. They appealed for a com-

prehensive tobacco control strategy resting on three basic

pillars: (1) an advertising ban, (2) restrictions on smoking

in public places, and (3) rapidly increasing tobacco taxes.

They pointed out that international evidence had shown

that increasing the excise tax on tobacco is the most effec-

tive tobacco control measure. Despite the fact that tobac-

co is addictive, numerous studies, performed in a variety

of countries, have shown that excise-induced increases in

tobacco prices causes tobacco consumption to decrease.

They also pointed out that international experience had

clearly shown that increasing tobacco excise taxes also

increases government revenue.

The tobacco control lobby wanted the government to ear-

mark a proportion of the tobacco excise taxes for general

health promotion strategies. However, the lobby groups

were unsuccessful with these requests, even after the

changes of the 1990s were implemented. As a rule, the

South African government does not earmark revenues.4 It

is argued that earmarking distorts the prioritization of gov-

ernment policies, and could lead to economic inefficiency

in the spending of these funds.

The MRC and NCAS regularly pointed out that the real

excise tax rate had been decreasing during the 1970s and,

even faster, during the 1980s. Even though this point

was well taken in the Department of Health, the Ministry

of Finance did not increase the tax. The government

explained its inaction as follows: (1) increasing the tax

would stimulate smuggling, and (2) an increase in the tax

might, in fact, decrease government revenue, because the

4 Exceptions to the rule are the Unemployment Insurance

Fund, the Skills Levy, and levies to fund regulatory bodies

of specific industries

6


7


tax-induced price increase would cause a sharp reduction

in demand. In providing these explanations, the Ministry

of Finance apparently did not question their empirical

foundations; they were generally taken as an article of

faith from the tobacco industry.

In the early 1990s, after the ban on the ANC had been

rescinded,5 and negotiations for a democratic transition

were taking place, the tobacco control groups started

lobbying the ANC for stricter tobacco control measures,

including rapid tax increases.6 The tobacco control lobby-

ists found an ally in Dr Nkozasana Zuma, the later Minister

of Health. She was passionately against smoking and, in

this regard, had the full support of the president-to-be,

Nelson Mandela.

In June 1994, less than two months after the democratic

transition, the Minister of Finance announced that the

government would increase the tax on tobacco products

to 50 per cent of the retail price. The Ministry of Health

and tobacco control lobby groups had been lobbying the

Ministry of Finance for a doubling of the excise tax that

year. Because of pressure exerted by the tobacco industry,

the Minister of Finance increased the tax by only 25 per

cent in 1994. Although this was a temporary setback for

the health community, the tax was increased by substan-

tially greater percentages in subsequent years. The guiding

principle for each of the subsequent tax increases was the

“50 per cent goal” announced in 1994.

The industry was naturally furious about the excise tax

increases. The Tobacco Institute of South Africa was par-

ticularly vocal about the “discriminatory” tax increases.

They argued that tobacco was already the most highly

taxed consumer product, and that such large tax increases

would encourage smuggling. Furthermore, they argued

that the tax would decrease tobacco consumption, which

would cause large numbers of workers to be retrenched.

The chairman of the Rembrandt Group, the country’s

largest cigarette manufacturer, wrote an open letter to

the Minister of Health in 1996 in which he argued that

smuggling was out of control, and that the government

was losing revenue as a result. He quoted the example of

Canada where smuggling had reached epidemic propor-

tions, which was reduced significantly after the taxes were

reduced.7

In 1996 the tobacco control lobby was strengthened when

the Economics of Tobacco Control (ETC) Project was

established at the University of Cape Town. Among other

things, the Project quantified how much revenue the gov-

ernment had lost during the 1970s and 1980s by allow-

ing the real excise tax to fall so sharply. This effectively

destroyed the industry’s argument that the government

might lose revenue by increasing the tax, because of the

reduction in consumption that it would cause. The tobacco

control lobby used these and other research results of the

ETC Project to counter the industry’s claims that tobacco

benefits the economy as a whole.8

The tobacco control lobby was heavily dependent on infor-

mation and news in order to retain the attention of the

public and the policymakers. Research results and tobacco

related news from developed countries certainly maintained

public awareness, but locally produced research results

generally received more media attention. The tobacco con-

trol lobby in South Africa used locally generated research

outputs to influence policymakers. This is important

because policymakers want to know what the impact of

certain interventions is likely to be on the South African

situation. They are generally not very interested in, or per-

suaded by, research that has been performed in a different

country, possibly under very different circumstances.

5 The African National Congress is a political party which

was founded in 1912. It was banned for 30 years under

the apartheid regime, from 1960 to 1990.

6 It must be noted that the National Party government

passed the country’s first tobacco control legislation in

1993. This was the result of persistent lobbying with the

then Minister of Health. The legislation was mild, even by

1993 standards, but it nevertheless represented a schism

between the NP government and the tobacco indus-

try. This legislation did not make any provisions for tax

increases. However, the more comprehensive legislation of

1999 did not include such provisions either.

7 However, subsequently it was found that the tobacco

industry was involved in the smuggling network. Litigation

is currently being brought against Brown & Williamson

regarding their role in the smuggling.

8 The ETC Project was not the first to investigate the eco-

nomic impact of smoking, but it was the most compre-

hensive. So, for example, a cost benefit analysis performed

by the University of Cape Town’s Health Economics Unit

in 1988 indicated that, for every R1 received in tobacco

taxes, the economy incurred medical costs and lost pro-

ductivity of R4.

8


9


The public’s reaction to the tax increases has been mixed.

In the “letters” section of newspapers people have

expressed both support for and disappointment in the tax

increases. Surveys indicate that most people, mainly non-

smokers, but also a sizeable proportion of smokers, gener-

ally support strategies aimed at reducing smoking.

Success of the Intervention

Internationally, tobacco control advocates generally

propose a comprehensive strategy in the fight against

tobacco. Such a strategy would typically consist of an

advertising ban, clean indoor air policies, restrictions on

sales to minors, an effective education programme, and

tax increases. The international literature indicates that, of

all these interventions, increases in tobacco taxes are the

most effective in reducing tobacco use.

The South African experience confirms these findings.

Econometric studies have shown that the average price

elasticity of cigarettes in South Africa is between 0.5

and 0.7. This means that, all other factors (e.g. income)

remaining constant, the consumption of cigarettes

decreases by between 5 and 7 per cent for every 10 per

cent increase in the real price of cigarettes.

In Figure 1 the relationship between cigarette consumption

and the real price of cigarettes is shown for the past four

decades. The figure clearly illustrates the inverse relation-

ship between these two variables. The increase in excise

taxes explains about half of the real price increase since

1991; the other half is attributed to the industry’s pricing

strategy (discussed in section 5). Since 1991 total cigarette

consumption has decreased by a third; in per capita terms

it has decreased by more than 40 per cent.

It was found that approximately 40 per cent of the

decrease in cigarette consumption was to be ascribed to

people giving up smoking. This is reflected in the fact

that the smoking prevalence percentage among adults

decreased from 33 per cent in the early 1990s to 27 per

cent in 2001. The other 60 per cent of the decrease in cig-

arette consumption is explained by the fact that smokers

are smoking less. In fact, average cigarette consumption

per smoker has decreased by approximately 20 per cent in

the past decade.

An analysis of smoking prevalence in South Africa reveals

that young people, low-income earners, black South

Africans and males have experienced the largest reductions

in cigarette smoking. Smoking prevalence among young

people decreased from 23 per cent in 1993 to 19 per cent

in 2000; among low-income earners from 31 per cent to

25 per cent; among black South Africans from 28 per cent

to 23 per cent, and among males from 51 per cent to 44

per cent. Surprisingly, smoking prevalence among black

South Africans has decreased despite a heavy tobacco

advertising campaign, specifically focused on emerging

0

100

200

300

400

500

600

700

1961

1963

1965

1967

1969

1971

1973

1975

1977

1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999

2001

Real retail price of cigarettes (cents/pack of 20; constant 1995 prices)

0

500

1000

1500

2000

2500

Consumption of cigarettes (millions of packs)

Rea

l ret

ail p

rice

of

ciga

rett

es

(cen

ts/p

ack

of 2

0; c

onst

ant

1995

pri

ces)

Con

sum

ptio

n of

cig

aret

tes

(mill

ions

of

pack

s)

Figure 1 Cigarette consumption and real prices of cigarettes in South Africa, 1961 to 2001

8


9


middle-class black South Africans in the second half of

the 1990s. The demographic groups that have not experi-

enced significant decreases in smoking prevalence include

females (although, admittedly, their smoking prevalence

level, at 13 per cent, is relatively low), high-income earners

(32 per cent), and white South Africans (36 per cent).

Studies performed in other countries indicate that young

people and the poor are more responsive to cigarette price

changes than older and more affluent people. The reason

is straightforward: an increase in the price of cigarettes

makes the product too expensive to those groups, with

the result that they either reduce their consumption or quit

altogether. The evidence from South Africa supports the

hypothesis that young people and the poor tend to reduce

their cigarette consumption by a greater percentage than

other groups in reaction to a price increase.

A related issue concerns the regressivity of an excise tax

on cigarettes. Some people are against using excise tax

increases as a tobacco control tool because it could have

a detrimental impact on the poor. Since the poor, vis-à-vis

the rich, generally spend a larger portion of their income

on tobacco products, they pay proportionally more tax.

This implies that the tax is regressive, which is regarded as

socially inequitable.

However, it has been shown that, in South Africa, an

increase in the tax on cigarettes causes a larger reduction

in cigarette consumption among the poor than among the

rich. This means that, while the absolute burden of the

tax is likely to increase for all income groups, the burden

on the poor, relative to that of the rich, is reduced. Thus,

even though excise taxes are regressive, increases in excise

tax reduce the regressivity of excise tax.

Other Effects of the Intervention

While the primary aim of a tobacco control strategy is to

reduce tobacco consumption, an agreeable by-product

of increasing the excise tax on tobacco is that it increases

government revenue. Column (j) of Table 1 shows that,

despite a 33 per cent reduction in tobacco consumption

over the past decade, real government revenue has more

than doubled. Since 1994, for every 10 per cent increase

in real excise tax, real excise revenues have increased by

approximately 6 per cent.

The tobacco industry has been ferocious in its opposi-

tion to any tobacco control measures, including excise tax

increases. Under the present government, the policy on

tobacco and tobacco control is unlikely to change. The

industry has had to drastically change its marketing strat-

egy under these difficult external conditions.

Whereas the pricing strategy of the cigarette manufac-

turing industry before the 1990s was focused primarily

on the growth in cigarette quantities, there is currently a

much stronger focus on the growth in price. Column (i)

in Table 1 shows that the real industry price (i.e. the retail

price less all taxes) did not change much between the

1960s and the early 1990s. If anything, the real industry

price decreased over this period. However, there has been

a very rapid increase in the real industry price since 1991

and especially since 1996. In 2001 the real industry price

of cigarettes had increased by more than 60 per cent

compared to the early 1990s. An analysis of the industry’s

major cost factors indicates that this increase is not the

result of an increase in the real costs of producing ciga-

rettes.

There is only one explanation: the industry is maintain-

ing its overall profitability by increasing the profit per

cigarette, despite the fact that quantities are falling. The

external environment has turned against the industry to

the extent that future growth in cigarette quantities seems

unlikely. Since the merger between Rothmans (of the

Rembrandt Group) and British American Tobacco (BAT)

in 1999, one company has controlled 95 per cent of the

South African cigarette market. This gives the newly cre-

ated BAT the necessary monopoly power to raise cigarette

9 The Department of Customs and Excise has recently com-

missioned a study aimed at quantifying the number of

smuggled cigarettes. However, to the author’s knowledge,

the results of this research are not yet known. An analysis

by the Economics of Tobacco Control Project, based on

rather cursory data, suggests that between 5 and 7 per

cent of cigarettes consumed in South Africa are not taxed

by the authorities. This percentage compares well with

most European countries.

10 In a recent newspaper article, BAT claims that 148 mil-

lion cigarettes were confiscated by the Department of

Customs and Excise in 2001. This is about 0.6 per cent

of total cigarette consumption in South Africa. However,

the proportion of smuggled cigarettes impounded by the

authorities is unknown. On the other hand, the smuggling

and trade in hard drugs (especially heroin and cocaine) is a

serious problem in South Africa, and attracts much media

attention.

10


11


prices above competitive levels. The industry can disguise

the retail price increases behind the well-publicised tax

increases.

A simulation analysis has indicated that this strategy has

been very beneficial to the cigarette manufacturing indus-

try. The strategy has increased total sales revenue for the

industry, with the result that the profitability of the indus-

try has been enhanced. Also, the strategy has reduced the

government’s ability to increase its excise tax revenue. By

increasing the real industry price the industry siphoned off

the extra revenue to itself, at the expense of the govern-

ment. The downside, from the industry’s perspective, is

that its pricing strategy has further reduced cigarette con-

sumption. The actions of the industry suggest that they

are in an end-game scenario, looking to milk the cow for

all it is worth before it finally dies.

From a tobacco control perspective the industry’s pric-

ing strategy has been beneficial, because it has reduced

cigarette consumption by a much greater percentage than

what the excise increases would have achieved in isolation.

It is ironic that the industry itself, in its attempt to further

its own short-term interests, followed a strategy that ben-

efited both the industry and the tobacco control lobby.

An issue of considerable importance in many countries

is that of cigarette smuggling. In South Africa, when-

ever the excise tax on tobacco products is increased, the

industry claims that this will increase smuggling activities.

Unfortunately, given the dishonest character of cigarette

smuggling, accurate data do not exist.9 Over the past

decade there have been very few reports of smuggled

or counterfeit cigarettes being impounded by the South

African police or customs authorities.10 While this is not

meant to imply that cigarette smuggling is not a problem,

it can be said with confidence that South Africa does not

experience the cigarette smuggling problems currently

experienced by the UK and, in previous years, by Canada.

Also, considering the official consumption statistics (as

shown in column (b) of Table 1), the decrease in cigarette

consumption in the past decade seems reasonable in view

of the very sharp increase in the real price of cigarettes.

An interesting characteristic of the South African cigarette

market is the absence of the Marlboro brand. In fact,

Philip Morris has no presence in South Africa. In a court

case in 1998 Rembrandt accused Philip Morris of smug-

gling cigarettes into South Africa via neighbouring coun-

tries. Tobacco industry documents clearly reveal that ciga-

rette smuggling is used to gain market entry and/or share

and to undermine tax policies. If smuggling, especially of

Marlboro, into South Africa becomes uncontrolled, this

could enable Philip Morris to formally enter the market.

For this reason it is in the interest of the dominant incum-

bent (first Rembrandt, and now BAT) to contain smug-

gling.

Apart from keeping Philip Morris out, the industry has

obvious motives for emphasizing the smuggling problem

in South Africa. If cigarettes were smuggled on a large

scale, the logical step, according to the industry, would

be to reduce the tax on cigarettes. This is exactly what

happened in Canada in the early 1990s.11 While some

informal bootlegging and some more organized smuggling

definitely occur, cigarette smuggling is not a serious threat

to the government’s excise tax policy.

Conclusion

South Africa has been able to significantly reduce its

tobacco consumption in a decade. While strong tobacco

control legislation and changing social norms have created

an environment where smoking is increasingly regarded

as socially unacceptable, the instrument with the biggest

impact has been excise taxation.

In South Africa, as in most countries, it is administra-

tively easy to change the excise tax on tobacco. What is

required is the political will to challenge the vested inter-

ests of the tobacco industry. In South Africa the Minister

of Health and nongovernmental organizations played a

pivotal role in implementing a comprehensive tobacco

control strategy, of which large increases in the excise tax

are a key part.

The effects of an increase in the excise tax on cigarettes

are soon felt: cigarette consumption decreases and gov-

ernment revenue increases. In South Africa the impact

of the excise tax increases was enhanced by the industry

when it increased the real retail price by more than the

increase in the real excise tax.

An important proviso concerns cigarette smuggling. While

South Africa’s experience can, in principle, be easily dupli-

cated in other countries, the success of such a strategy

will depend crucially on whether the country can contain

11 However, the logic is flawed. Evidence from several coun-

tries shows that tax reductions do not, in themselves,

reduce smuggling.

10


11


cigarette smuggling within reasonable limits. In South

Africa cigarette smuggling certainly did not undermine

the strategy, despite the industry’s claims to the contrary.

Although individual countries may differ, international

experience shows that, despite smuggling, higher tobacco

tax decreases tobacco consumption and increases govern-

ment revenue. Smuggling erodes but does not completely

destroy the benefits of higher taxes.

Regarding cigarette smuggling, the industry has its own

reasons for exaggerating the threat. In countries where

smuggling could be a problem, the authorities should

impose strong control mechanisms, including stiff penalties,

cooperative efforts with customs and law enforcement offi-

cials in other countries, and laws to make exporters respon-

sible for their exports all the way to a final legal and taxed

destination, thus discouraging potential smugglers.

Tobacco Excise Taxation in South Africa - · PDF fileannounced that it would accelerate the...

Documents

Transcript of Tobacco Excise Taxation in South Africa - · PDF fileannounced that it would accelerate the...